Ultrasonics 70 (2016) 275–283

Contents lists available at ScienceDirect

Ultrasonics

journal homepage: www.elsevier .com/locate /ul t ras

Investigation of polymer-shelled microbubble motionsin acoustophoresis

http://dx.doi.org/10.1016/j.ultras.2016.05.0160041-624X/� 2016 Elsevier B.V. All rights reserved.

⇑ Corresponding author at: Department of Medical Engineering, School ofTechnology and Health, KTH Royal Institute of Technology, Alfred Nobels allé 10,SE-142 51 Stockholm, Sweden.

E-mail address: dmitryg@kth.se (D. Grishenkov).

Satya V.V.N. Kothapalli a, Martin Wiklund b, Birgitta Janerot-Sjoberg a,d,e, Gaio Paradossi c,Dmitry Grishenkov a,d,e,⇑aDepartment of Medical Engineering, School of Technology and Health, KTH Royal Institute of Technology, SE-142 51 Stockholm, SwedenbDepartment of Applied Physics, KTH—Royal Institute of Technology, SE-106 91 Stockholm, SwedencDipartimento di Chimica, Università di Roma Tor Vergata, 00133 Rome, ItalydDepartment of Clinical Science, Intervention and Technology, Karolinska Institute, SE-142 51 Stockholm, SwedeneDepartment of Clinical Physiology, Karolinska University Hospital, SE-142 51 Stockholm, Sweden

a r t i c l e i n f o

Article history:Received 10 June 2015Received in revised form 30 March 2016Accepted 19 May 2016Available online 1 June 2016

Keywords:AcoustophoresisUltrasound contrast agentRadiation forceUltrasound standing waveAcoustic contrast factor

a b s t r a c t

The objective of this paper is to explore the trajectory motion of microsize (typically smaller than a redblood cell) encapsulated polymer-shelled gas bubbles propelled by radiation force in an acousticstanding-wave field and to compare the corresponding movements of solid polymer microbeads. Theexperimental setup consists of a microfluidic chip coupled to a piezoelectric crystal (PZT) with a reso-nance frequency of about 2.8 MHz. The microfluidic channel consists of a rectangular chamber with awidth, w, corresponding to one wavelength of the ultrasound standing wave. It creates one full waveultrasound of a standing-wave pattern with two pressure nodes at w/4 and 3w/4 and three antinodesat 0, w/2, and w. The peak-to-peak amplitude of the electrical potential over the PZT was varied between1 and 10 V. The study is limited to no-flow condition. From Gor’kov’s potential equation, the acoustic con-trast factor,U, for the polymer-shelled microbubbles was calculated to about �60.7. Experimental resultsdemonstrate that the polymer-shelled microbubbles are translated and accumulated at the pressureantinode planes. This trajectory motion of polymer-shelled microbubbles toward the pressure antinodeplane is similar to what has been described for other acoustic contrast particles with a negative U.First, primary radiation forces dragged the polymer-shelled microbubbles into proximity with each otherat the pressure antinode planes. Then, primary and secondary radiation forces caused them to quicklyaggregate at different spots along the channel. The relocation time for polymer-shelled microbubbleswas 40 times shorter than that for polymer microbeads, and in contrast to polymer microbeads, thepolymer-shelled microbubbles were actuated even at driving voltages (proportional to radiation forces)as low as 1 V. In short, the polymer-shelled microbubbles demonstrate the behavior attributed to thenegative acoustic contrast factor particles and thus can be trapped at the antinode plane and thereby sep-arated from particles having a positive acoustic contrast factor, such as for example solid particles andcells. This phenomenon could be utilized in exploring future applications, such as bioassay, bioaffinity,and cell interaction studies in vitro in a well-controlled environment.

� 2016 Elsevier B.V. All rights reserved.

1. Introduction

In a current clinical practice a suspension of micro-size gascore bubbles is used as and efficient ultrasound contrast agent(UCA) [1]. As early as 1968 Gramiak and Shah [2] demonstratedthe enhancement of the image contrast in echocardiographic

study of aorta following the injection of the suspension of agi-tated saline solution containing free gas bubbles. In order toincrease stability of the free gas bubbles, otherwise dissolvingin a surrounding media in fractions of a second, the gas corewere encapsulated within the solid shell. The solid encapsulatingshell is typically made of albumin protein [3], phospholipids [4],or polymers [5]. Moreover, the outermost surface of the encapsu-lating shell can be further modified to accommodate ligands,pharmacological molecules or genes for molecular imaging andlocalized drug delivery [6–8]. In these applications the acousticradiation force which propels and accumulates the

276 S.V.V.N. Kothapalli et al. / Ultrasonics 70 (2016) 275–283

free-circulating in blood stream bubbles on the target is of partic-ular interest [9,10].

For more than a century, researchers have been exploring theeffects of acoustic radiation force on particles suspended in liquidmedia and the particles’ motion with the acoustic waves [11]. Inthe last two decades, several medical innovations have been basedon this acoustic radiation-force phenomenon, including vibro-acoustography [12], shear wave elastography [13], acousticradiation-force impulse imaging [14], magnetic resonance acousticradiation-force imaging [15], the assessment of the viscoelasticproperties of tissue [16], and the precise manipulation of cells orparticles in a standing wave [17]. This study focuses on the appli-cation of acoustic radiation forces acting on microbubbles instanding-wave (stationary sound fields) acoustic fields. Blake [18]reported that millimeter-size gas bubbles (resonance frequencywas well below the excitation frequency) suspended in a liquidwere drawn to the pressure antinodes by the radiation force in astanding wave. The bubbles coalesced owing to the secondary radi-ation effect (also known as Bjerknes force) when they moved closeto each other and merged into a larger bubble with a lower reso-nance frequency. Once the resonance frequency of a resulting bub-ble was comparable to the driving frequency, then it propelledaway from the pressure antinode to the pressure node [19]. Apartfrom the frequency effect, acoustic pressure also plays significantroles in bubble motion. At a high enough acoustic pressure, eventhe driving frequency is well below the bubbles’ resonance fre-quency, resulting in bubble translation and precision around thepressure node [20]. On the other hand, Kundt and Lehman demon-strated that when radiation force was exerted on solid Styrofoamchips, particles accumulated at the pressure nodes in acousticstanding-wave fields [21]. The motion of particles and their direc-tions in a standing-wave acoustic field depend on several factors,such as driving frequency, size, density, and compressibility ofthe particle [22,23]. If the particle possesses greater compressibil-ity and lower density than the surrounding liquid media, it tends tomove toward the pressure antinode; otherwise, the particle movestoward the pressure node [22,23]. Note that this phenomenon isapplicable only when the resonance frequency and size of the par-ticle are well below the driving frequency and the wavelength ofthe incidental acoustic wave.

Based on particle responses upon the radiation force in acousticstanding-wave fields, the field of acoustophoresis (acoustic manip-ulation) has emerged. Acoustophoresis is an emerging clinical tool,especially in noncontact cell handling, useful for concentrating,sedimenting, sorting, and purifying [24–26]. Laurell et al. [27] suc-cessfully separated lipid particles from blood cells by using theacoustophoresis separation method. Analogously, the successfulseparation of platelets and serum from blood [28], the separationof lipid particles frommilk [29], the separation of crude oil dropletsfrom environmental water samples [30], and the separation of tar-geted cells with biofunctional elastomeric particles from nontar-geted cells [31,32] have all been reported. Recently, tunableglass-shelled core (core: air, water, and steel) particles were alsoshown as negative acoustic contrast particles [33]. In that laterstudy, the authors mentioned that the glass-shelled core-particlescenario can extend to UCAs but did not take this into account.Moreover, the authors did not mention the particular type ofUCA that is, whether thin- or thick-shelled air-filled gas bubbleswere used—nor did they report the type of shell material, such aslipids or polymers.

Thick- or thin-shelled microbubbles are defined based on theratio between shell thicknesses and the total microbubble radius,where bubbles with a ratio below 5% are considered thin [34]. Inthe interest of studying sonoporation or cell lysis, Khanna et al.[35] introduced thin Albunex-shelled (human-protein-shelled)microbubbles (Optison�) and erythrocytes (red blood cells) simul-

taneously into ultrasound standing-wave fields at a frequency ofabout 1.5 MHz. The authors reported that in the presence of Opti-son�, the erythrocytes moved more vigorously and randomly inthe acoustic field and released significant amounts of hemoglobin.However, the Optison� microbubbles also disappeared within thefirst frame—that is, within a few milliseconds. In the current study,we utilized thick-shelled polymer microbubbles with the ratio ofshell thickness to microbubble radius above 5% [36]. Comparedto the shell of lipid- or protein-shelled microbubbles [37,38], thethick encapsulated polymer shell offers increased mechanical sta-bility resulting not only in shelf-life of several months but also inextended circulation time during in vivo tests. Thick shell alsooffers larger volume for incorporation of therapeutical gas [39] orpharmacological relevant molecules [39] to be delivered locallyfollowing ultrasound excitation.

The aim of the study was to investigate the movements ofpolymer-shelled air-core microbubbles induced by radiation forcein acoustic standing-wave fields and to compare these with thecorresponding movements of solid polymer beads, currently usedas a blood mimicking phantom.

Prior to the experiments, we estimated the acoustic contrastfactor, U, value of our polymer-shelled microbubbles suspendedin water following the Gor’kov potential theory [22]. The sign Upredicts the particle trajectory motion, which is extensivelydescribed in the theory section. Furthermore, the response of thepolymer-shelled microbubbles at different driving voltages acrossthe PZT is explored so as to identify the relation between acousticpressure and the trajectory motion of the microbubbles. Our exper-imental results are compared with those presented in well-established studies of polymer microbeads [40,41]. Finally, thepaper concludes with a thorough discussion of the fundamentalphysical principles behind the observed phenomena and notespotential applications for polymer-shelled microbubbles.

2. Theory

2.1. Radiation force

Suspended particles in a liquid experience both axial and trans-verse acoustic radiation forces when they are subjected to standingwave acoustic fields. The axial radiation force acts toward thedirection of the wave propagation, which is responsible for a driv-ing particle to either pressure node (velocity antinode) or pressureanti-node (velocity node). The transverse radiation force is actingperpendicular to the wave propagation, that is accountable forgrouping the particles to clusters. The mathematical representa-tion of the primary axial radiation force [22] is given in Eq. (1),

Frad ¼ �Vf 12bsrhp2

1i �3f 24qsrhv2

1i� �

; ð1Þ

where V is the volume of the particle; f 1 ¼ 1� bpbsand f 2 ¼ 2ðqp�qsÞ

2qpþqsare

the monopole and dipole scattering coefficients; qp and qs and bpand bs are the densities and compressibility of the particle andthe surrounding media, respectively; p1 is the incidental pressure;and v1 is the velocity of the particle. In 2-D acoustophoresis,microparticles move in a horizontal direction, x(t), and in atransversal direction, y(t). The acoustic pressure field in transversemotion is p1 = pa cos(ky), and when this is substituted in Eq. (1),the radiation force acting on the transverse field becomes [42]

Frady ¼ 4p

3kya3EacUðb;qÞ sinð2kyyÞ; ð2Þ

where k(=2p/k) is wavenumber; k is the wavelength equal to chan-nel width,w; a is the radius of the single spherical particle; Eac is the

S.V.V.N. Kothapalli et al. / Ultrasonics 70 (2016) 275–283 277

time-averaged acoustic energy density; and U(b, q) is the acousticcontrast factor, which equals 1

3 f 1 þ 12 f 2.

2.2. Acoustic contrast factor

The direction of particle movement depends on the sign of theacoustic contrast factor,U. As shown in Fig. 1, particles with a pos-itive U are translated and focused at the pressure node (i.e., veloc-ity antinode). Particles with a negative U move in the oppositedirection and are trapped at pressure antinode (velocity node).

Barnkob et al. [40] estimated the U of polyamide microbeadssuspended in water to about +0.24. Thus, microbeads are draggedinto pressure nodes by the radiation force in a standing waveacoustic field. The polymer-shelled gas particle/bubble however,is a combination of a gas core and a complex polymer shell. Thus,the density is calculated by considering the cumulative sum of aircore and polymer shell densities, which can be expressed asqp = qshell(1 � a3) + qgasa3. Here the parameter a is the ratiobetween inner, R01 and outer, R02, radii of the polymer-shelledgas particle, and equals 0.84. Fernandes et al. [43] reported that,the encapsulated PVA shell is the composition of 80% water and20% PVA moieties. The density of PVA and water is about1269 kg/m3 and 1000 kg/m3 respectively and the total shelldensity thereby equals 1053 kg/m3. Thus the total density of thepolymer-shelled gas particle, qp, is 429.6 kg/m3.

A pulsating microbubble in lower acoustic pressure can bemodelled as a damped linear harmonic oscillator. Grishenkovet al. [36] derived the linearized version of the non-linear equationof motion of polymer-shelled microbubbles given in Eq. (3), whichis similar to the damped harmonic oscillator model as given inEq. (4),

�x2 þ j 4cqSR

201R

302

R301lLxþ VSG

00ðxÞ� �h

þ 3cqSR

201

jPG;eq þ 4VS

3R302G0ðxÞ

� �iXðxÞ ¼ DP̂

cqSR201;

ð3Þ

�x2 þ 2jdxþx20

� �XðxÞ ¼ DP̂

cqSR201

; ð4Þ

where R01 and R02 are the inner and outer radii of the microbubbleat rest; lL is the shear viscosity in surrounding liquid; p0 is theambient pressure; pac is an incidental acoustic pressure;VS ¼ R3

02 � R301 ¼ R3

2 � R31 describes the shell incompressibility

assumption in the model; PG,eq represents the gas pressure at equi-librium that is assumed to be equal to p0; G0(x) and G00(x) are thefrequency-dependent storage and loss modulus; qs and qL are thedensities of shell and liquid; x(=2pf) and x0(=2pf0) are the parti-

Fig. 1. Illustration of the acoustic pressure standing-wave profile and particlemotion depending on different acoustic contrast factors, U, in a liquid suspension.

cle’s driving frequency and resonance frequency, respectively; andj is the polytrophic gas constant that varies between 1 (isothermal)and 1.4 (adiabatic).

Comparing Eqs. (4) and (5) allows us to identify the mass

(m ¼ cqSR201), friction R ¼ 2ðR3

01lLxþ VSG00ðxÞÞ=R3

02

� �, and stiff-

ness KP ¼ jPG;eq þ 43R302

VSG0ðxÞ

� �. Here, the storage modulus G0(x)

can be expressed as G0(x) = Geq + G1x3/4, where Geq is the equilib-rium shear modulus and G1 is the frequency-dependent shearmodulus. Both Geq and G1 were identified by fitting theoreticalattenuation spectra with measured attenuation spectra, whichwere reported as 10.5 MPa and 5.5 Pa/(rad/s)3/4, respectively [44].Herein the driving frequency, f, of PZT equals 2.8 MHz; therefore,G0 is 11.9 MPa. The mean external radius of the bubble is 1.9 lm,and PG,eq is the atmospheric pressure—that is, about 105 Pa. Theresultant Kp is equal to 12.2 MPa, and the inverse of Kp, thecompressibility, bp, is about 0.08 MPa�1. Further, the monopole scat-tering coefficient or compressibility factor, f1, would be 181.4, andthe dipole scattering coefficient or density factor, f2, would be +0.4.Thus, the acoustic contrast factor, or U value, is estimated to �60.7.

2.3. Transverse particle path, acoustic energy density and localpressure amplitude

The trajectory motion of the single particle, either microbubbleor microbead, in acoustophoresis can be assessed using the equa-tion for transverse particle part y(t) where the expression for trans-verse radiation force (Eq. (2)) is balanced with the Stokes drag force[42]

yðtÞ ¼ 1ky

arctan tan½kyyð0Þ� exp 4U9g

ðkyaÞ2Eact� ��

; ð5Þ

where y(0) is the initial position of the particle at t = 0; and g is theviscosity of the surrounding liquid.

Inverting the Eq. (5) lead to expression for the acoustic energydensity.

Eac ¼ 9g4UðkyaÞ2t

lntan½kyyðtÞ�tan½kyyð0Þ�

� �: ð6Þ

Keeping in mind that both structural and mechanical character-istics of the microbeads are known [40] it is possible using theexperimentally measured accumulation time, t, to estimate thevalue for acoustic energy density at different driving voltage. Fur-thermore the pressure amplitude, pa, in a chip can be estimatedfrom the relation pa ¼ 2

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiqsc2s Eac

p; where qs and cs are the density

and speed of sound of surrounding liquid.Thus the indirect calibration of the acoustic standing-wave field

can be achieved.

3. Material and method

3.1. Particles

In this research, two particle types were utilized namely, poly-mer microbeads and polymer-shelled microbubbles. The testifiedpolymer microbeads’ (polymer was made of polyamide molecules)suspension was prepared by adding a mixture of Milli-Q water(with 0.01% Tween20) to blood-mimicking fluid (EU-DFS-BMF-ver.1 for flow Doppler Phantoms, Danish Phantom Design, Den-mark) in a ratio of 1:9. Barnkob et al. [45] reported that the concen-tration of microbeads suspension was about 3.5 � 1011 ml�1, whileits diameter was 4.5 ± 0.7 lm. The polymer-shelled microbubblesresulted from the encapsulation of microsize gas bubbles with aPVA (poly vinyl alcohol), hydrogel. The polymer-shelled microbub-bles were reproduced in our lab following the protocol developed

278 S.V.V.N. Kothapalli et al. / Ultrasonics 70 (2016) 275–283

by Cavalieri et al. [37]. Concisely, the PVA powder (Sigma–Aldrich,Chemie GmbH, Germany) was dissolved in distilled water, and theresulting solution was mixed with sodium (meta)periodate at atemperature of 80 �C. Later, this solution was subjected to ahomogenizer ULTRA-Turrax� (Ika, Germany) for two hours. Duringthe high-stirring process, the polymer moieties with hydrophilicheads and hydrophobic tails were rearranged at the water/airinterface. As a result, stabilized gas particles were obtained. Thediameter and concentration of polymer-shelled microbubblesequal 3.8 ± 0.6 lm and 5 � 108 ml�1, respectively. The thicknessof the polymer shell reported by Poehlmann et al. [44] was approx-imately 300 nm. Thus, the ratio of shell thickness to microbubbleradius is 16%.

3.2. Experimental set-up

The schematic representation of the experimental setup is pre-sented in Fig. 2. A microfluidic chip (GeSim GmbH, Dresden, Ger-many) made of a glass–silicon-glass composite structure [45]was placed under the transmission microscope (Axiovert 40 CFL,Carl Zeiss, USA). The microchip included the microfluidic channeland a rectangular chamber. The depth and width of the channelwere about 110 � 300 lm2, and the length and width of the rect-angular chamber were about 8900 � 530 lm2. The microchip hadone inlet and one outlet. A PZT was coupled to the upper surfaceof the microchip using conductive adhesive gel (Tensive�, Labora-tories, Inc., USA). An impedance analyzer (Model 16777k, Sine-Phase Instruments GmbH, Moedling, Austria) found thefundamental resonance frequency of the attached PZT to be2.8 MHz.

The PZT was driven to operate by a function generator (AFG3022, Tektronix Inc., USA) with a continuous sinusoidal wave.The width of the rectangular chamber was designed to equal onewavelength, k, of the ultrasound standing wave—that is, 535 lm.Thus, the cage acted as a full wave resonator, and the superpositionof vertical and horizontal waves in the full wave resonator resultedin two pressure node planes (at k/4 and 3k/4) and three pressureantinode planes (at 0, k/2, and k). The microscope, equipped witha 10� objective (NA = 0.25; Zeiss, Germany), was focused on therectangular chamber. Images were captured by a digital camera(SLT-A77V, 77a, Minato, Japan) with a frame rate of 50 Hz. In these

Fig. 2. Schematic representation of the experimental setup consisting of thesilicon/glass microchip connected to the PZT transducer mounted below themicroscope and the CCD camera. The detail (top-view) depicts the rectangularchamber and the acoustic standing-wave profile in the region of interest.

experiments, the concentrations of both microbeads and polymer-shelled microbubbles were diluted with distilled water to equal106 ml�1. The solution was introduced into the channel throughthe inlet using a syringe (BD Luer-LokTM Tip, Sweden). The flow ofthe particle suspension in the microchip was interrupted in orderto bring the solution into a stationary position, thereby avoidingpressure gradient and microstreaming effects. Thus, all experi-ments were performed in no-flow conditions.

3.3. Accumulation time

The excitation voltage at the PZT (which is indirectly propor-tional to the applied pressure on the particle) was incrementallyincreased from 1 to 10 V. The motion of the particles was recordedfor about two minutes at each voltage level and stored in a PC. Therecorded videos were post-processed, and the motion of the parti-cles was tracked in Matlab�. The image size is about 1080 � 1920,and particles in the image are visualized as black dots against alighter background. A thresholding method was implemented totrace particles. The particle intensity profile was constructed bysummation of all black dots along the channel width (the rows ofimage). The particle intensity profile demonstrates the number ofparticles in specific location along the width of the channel. Withan activation of the PZT, the particles moved toward and accumu-lated at either the node or the antinode, where the correspondingregion of interest (RoI) were selected. The particle intensityincreased with time and saturated once the particles were trappedat either node or antinode. The time from PZT activation to maxi-mum in particle intensity defined the particles’ accumulation time.

4. Results

4.1. Microbeads

As shown in Fig. 3a, the solid polymer microbeads were ran-domly distributed across the chip and settled on the bottom planeof the channel when they were not subjected to acoustic standing-wave fields. With an activation of voltage (about 9 Vpp) across thePZT, the primary acoustic radiation forces dragged the polymermicrobeads to the pressure node planes. As shown in Fig. 3b–f,the polymer microbeads aligned with the pressure node planesand formed two streamlines at k=4 and 3k=4 as time progressed.The distribution of microbeads across the rectangular chamberwas traced with respect to time. Fig. 4 presents the intensity pro-files of the polymer microbeads that correspond to the images inFig. 3. The intensity profiles clearly indicate that there are twopressure node planes, at about 130 lm (=k/4) and at 400 lm(=3k/4).

To calculate the accumulation time of microbeads, the region atthe pressure node (at about 130 ± 40 lm) is considered a region ofinterest (RoI) for tracking particles accumulation over time, asillustrated in Fig. 4. The maximum intensity value in the RoI withrespect to different time intervals is illustrated in Fig. 5. After acti-vation of the PZT, the concentration of microbeads graduallyincreased to the maximum. The particle intensity reached a maxi-mum, and the curve became saturated after the maximum. Thus,most of the polymer microbead population relocated at the pres-sure node plane at the point when the particle intensity reachedits maximum. The time from PZT activation to the maximum par-ticle intensity is the accumulation time, which is about 30 s for anelectrical potential of about 9 V across the PZT.

The knowledge of accumulation time at varied electrical poten-tial across the PZT is further employed for estimation of acousticstanding-wave field following Eq. (6) and microbeads characteris-tics reported in [40]. Both acoustic energy density, Eac, and local

Fig. 3. The microscopic images focused along the rectangular chamber at different time intervals when the polymer microbeads were (a) not subjected to an acousticstanding-wave field and (b–f) subjected to an acoustic standing-wave field. As time progressed, the particles shifted toward the pressure node planes. The particles appear asblack dots against a lighter background.

Fig. 5. Tracking the maximum intensity values around the ROI, as shown in (a)—that is, at one the pressure node plane at 133 ± 40 lm.

Fig. 4. Normalized intensity profiles of the polymer microbeads across therectangle chamber at different time intervals, corresponding to images in Fig. 3.

S.V.V.N. Kothapalli et al. / Ultrasonics 70 (2016) 275–283 279

pressure amplitude, Pa, are estimated at electrical potential, U,between 3 and 10 V and reported in Fig. 6. The acoustic pressureamplitude, Pa, as expected depends linearly on the driving voltage,U. The value R2 of the linear fit Pa(U) = 14.86U + 21.67 is equal to0.98. At the maximum voltage considered in this study the acousticenergy density is 3 J/m3 which correspond to the pressure ampli-tude of 165 kPa. These results are consistent with the previouslyreported values estimated using either individual particle trackingby Barnkob et al. [42], or light-intensity method by Barnkob et al.[40]. Worth mentioning is that at a pressure below 100 kPamicrobubbles behave as a linear scatterers whereas in a rangebetween approximately 100 kPa and 1 MPa polymer bubbles oscil-lation is nonlinear with second and higher order harmonics pre-sent in a spectra [46]. The fracturing of polymer microbubbleswere extensively characterized in [47] and occurs at a peak nega-tive pressure above 1 MPa at a frequency 2.2 MHz. As a result nofracturing, fragmentation or destruction of the microbubbles isexpected at a pressure values considered in this study.

4.2. Polymer-shelled microbubbles

Fig. 7 illustrates the polymer-shelled microbubbles positionsbefore and after they were subjected to a standing-wave acousticfield. Here the driving voltage across the PZT equaled 9 V. Afterthe PZT activation, the polymer-shelled microbubbles were trans-lated to antinode planes at the center (w/2) and near the walls (0and w) along the channel. Once the polymer-shelled microbubbleswere driven to the pressure antinode planes, they aggregated atdifferent locations transverse to the initial direction of the radia-tion force. The reason for this aggregation is because of the weakeraxial component of the 2D acoustic radiation force [42,48], but wealso believe that the secondary acoustic force accelerates theaccumulation. The intensity profiles of the polymer-shelledmicrobubbles’ distribution in the chamber at time intervals corre-sponding to the images in Fig. 6 are given in Fig. 8. At the edges,the polymer-shelled microbubbles appear blurred and are not clearly

2 3 4 5 6 7 8 9 10 110.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0

20

40

60

80

100

120

140

160

180Eac

Acousticenergydensity,E

ac[J/m

3 ]

Electrical potential,U, [Volts]

Pa

Linear fit

Pfita (U) = A + B*UA=21.67 6.00B=14.86 0.87R2 = 0.98

Pressure

amplitude,p

a[kPa]

Fig. 6. Calibration of the acoustic standing-wave field inside the chamber. Acousticenergy density (squares) and corresponding pressure amplitude (triangles) areplotted with respect to the electrical potential applied to the PZT crystal. Solid linerepresents the linear fit of the pressure amplitude as a function of the drivingvoltage.

280 S.V.V.N. Kothapalli et al. / Ultrasonics 70 (2016) 275–283

distinguishable from the wall background; for this reason, thepolymer-shelled microbubbles at the edges were not tracked. Asshown in Fig. 9, the RoI selected was 265 ± 40 lm (at k/2) in orderto measure the accumulation time of the polymer-shelledmicrobubbles.

The accumulation time of polymer microbeads and microbub-bles at the node and antinode planes, respectively, versus the pres-sure amplitude is shown in Fig. 10a and b. It is evident that theaccumulation time for both polymer-shelled microbubbles andpolymer microbeads decreases as the pressure increases. The pri-mary radiation forces generated by the PZT were too weak to drivethe microbeads to the pressure node planes at a pressure below60 kPa, which corresponds to the driving voltage less than 3 V. Atpressure 165 kPa (10 V), the polymer-shelled microbubbles weredriven to the antinode planes in less than half a second, whilemicrobeads took approximately 26 s.

4.3. Mixed solution

The mixed solution (composed of both polymer-shelledmicrobubbles and polymer microbeads) was introduced into the

Fig. 7. The microscopic images of polymer-shelled microbubbles (represented by blackshelled microbubbles are not subjected to an acoustic standing-wave field and (b–f) whe0.2, 0.4, 0.6, 0.8, and 1.0 s respectively.

microfluidic chip to attest the acoustophoresis separation betweenparticles with a positive acoustic contrast factor (polymer microbe-ads) and others with a negative acoustic contrast factor (polymer-shelled microbubbles). As shown in Fig. 11, the polymer-shelledmicrobubbles were collected and focused in the antinode plane(located in the lower plane), while microbeads were situated inthe pressure node plane (located in the upper plane) in the pres-ence of an ultrasound standing wave. The density of thepolymer-shelled microbubbles was half the density of water, theweight of the displaced water was greater than the weight of theparticle, and the buoyancy forces moved the polymer-shelledmicrobubbles to the upper surface of the chip. On the other hand,the density of microbeads was slightly greater than the density ofwater, and the weight of the displaced water was greater than theweight of the microbeads; thus, microbeads were deposited in thebottom plane of the chip. This is visualized clearly in Fig. 11b and c,where polymer microbeads and polymer-shelled microbubbles canbe seen concentrated in different planes along the z-axis.

5. Discussion

In this study, we examined the movements (directions andvelocities) of polymer-shelled microsize gas-core bubbles withrespect to radiation force in acoustic standing-wave fields. Polymermicrobeads (solid microparticles) were employed to mimic theacoustic behavior of red blood cells and also allow us to estimatethe acoustic energy density and peak pressure inside the standingwave resonator.

In contrast to solid microparticles, such as plain polymermicrobeads [42], the polymer-shelled microbubbles have shownthe tendency to be translated and trapped at the pressure antinodeplanes in ways similar to lipid (fat) particles [27], oil droplets [30],elastomeric particles [32], and hollow and glass-shelled core parti-cles [33]. Although the glass-shelled core particles’ movement canextend to the motion of UCAs, a scenario mentioned by Leibacheret al. [33], several other phenomena should be taken into accountfor UCAs, such as resonance frequency, damping, and compressibil-ity. Indeed, the compressibility of glass-shelled core particles is farless than the compressibility of UCAs.

According to the theory, for highly compressible lm-size gasbubbles suspended in water media, the acoustic contrast factor, U,has a huge negative value, about �5640. Thus, gas bubbles arenaturally negative acoustic contrast particles, and they certainly

dots) in a rectangular chamber at different time intervals. (a) When the polymer-n they are subjected to an acoustic ultrasound standing-wave field at an intervals of

Fig. 8. Normalized intensity profiles of the polymer-shelled microbubbles acrossthe rectangular chamber at different time intervals, corresponding to the imagesgiven in Fig. 6.

Fig. 9. Tracking the maximum intensity values around 265 ± 40 lm (i.e., aroundpressure antinode planes).

S.V.V.N. Kothapalli et al. / Ultrasonics 70 (2016) 275–283 281

have a tendency to move toward the pressure antinode under theradiation force in an ultrasound standingwave [18]. However, rapiddissolution of gas in water makes it more difficult to document thisphenomenon experimentally in unprotected microsize gas bubbles

60 80 100 120 140 160 180

20

40

60

80

100

120

140

160

180

200 Microbeads

Pressure amplitude, [kPa]

Accumulationt ime[s]

(a)

Fig. 10. The accumulation time of microbeads (a) at pressure node and microbub

(e.g., a three-micron-size free gas bubble diffused in a liquid within0.02 s [49]), and they are less useful for practical applications.

As mentioned in the introduction, thin-shelled microbubbles(Optison�) have also disappeared rapidly in high-intensityultrasound standing-wave fields during cell viability studies [35].In addition, the resonance frequencies of these UCAs varied withinthe 1–5 MHz range owing to differences in size distribution(poly-dispersion) [49]. Given excitationwithin this frequency range,the bigger bubbles (i.e. resonance frequency lower than excitationfrequency) move to the pressure node, while the smaller bubbles(i.e. resonance frequency higher than excitation frequency) moveto the pressure antinode [18]. It is therefore challenging to studythe polydispersed thin-shelled microbubbles in acoustophoresis.On the other hand, the polymer-shelled microbubbles utilized inthis study are relatively monodispersed in size and possess highermechanical stability than thin-shelled microbubbles do. Theresonance frequency of polymer-shelled microbubbles at about12 MHz [36] is far from the excitation frequency (2.8 MHz) gener-ated in this study by the PZT attached to the microchip, and thepolymer-shelled microbubbles did not exhibit any resonance effect(chaotic motion) during the experiments. Moreover, thesemicrobubbles remained intact for longer periods in the actuationfiled. In other words, the polymer-shelled microbubbles were notdestroyed at any pressure values considered in this study.

The acoustic contrast factor, U, for currently reported acousticcontrast particles (both negative and positive) such as blood cells,microbeads, lipid particles, oil droplets, and elastomer particlesfalls between �1 and 1. This is because the density differenceand the compressibility difference from the surrounding fluidmedia are not significant. For example, the monopole scatteringcoefficient (i.e., compressibility factor), f1, and the dipole scatteringcoefficient (i.e., density factor), f2, of an oil droplet are �0.44 and�0.07, respectively, and the resulting U is �0.18 [30]. In the caseof polymer-shelled microbubbles, the U is estimated to be �60.7,and it is several orders of magnitude greater than the U of cur-rently reported negative contrast particles, even if the U, owingto a thick stabilizing shell, is appreciably decreased relative to thatof free gas bubbles. The total density of a polymer-shelledmicrobubble, including its gas core, is about half the density ofwater. As a result, f2 becomes positive, and its value is equal+0.4. The polymer shell is compressible, but the compressibilityof the gas core is several magnitudes higher than that of the shell.This renders f1 negative, making its value about �181.4. Compress-ibility therefore plays a major role in allowing polymer-shelledmicrobubbles to behave as negative acoustic contrast particles.

Moreover, the polymer-shelled microbubbles respond morequickly to the acoustic radiation force than the polymermicrobeads do. They relocate at the pressure antinode much faster

60 80 100 120 140 160 1800

1

2

3

4

5(b)

Microbubbles

Pressure amplitude, [kPa]

Accumulationt ime[s]

bles (b) at pressure antinode plane with respect to the pressure amplitude.

Fig. 11. Microscopic images of mixed polymer-shelled microbubbles’ and solid polymer microbeads’ suspension in the microchip (a) before and (b and c) after beingsubjected to acoustic standing-wave fields. Owing to the density differences between bubbles and solid particles, they aligned in different planes. In (b) the bubbles are out offocus at pressure antinode planes, and the beads at pressure node planes are in focus, whereas in (c) the beads are out of focus at nodes, and the bubbles are in focus.

282 S.V.V.N. Kothapalli et al. / Ultrasonics 70 (2016) 275–283

(typically around 40 times faster) than microbeads do at the pres-sure node. Fig. 10a and b illustrates that the relocation timedecreases as the pressure amplitude increases. Both microbubbleand microbeads experience the same acoustic radiation pressureat the same driving voltage across the PZT; however, greater parti-cle velocity is obtained for the polymer-shelled microbubbles. Thiscan be explained by Eq. (1), which indicates that the radiationforce, F y

rad, acting on the particles is directly proportional to the vol-ume of the particle, V, the square of the pressure amplitude, p2, thefrequency, f, and the acoustic contrast factor, U. The volume of themicrobeads (5.23 � 10�16) is 2.2 times higher than the volume ofthe polymer-shelled microbubbles (2.29 � 10�16), while the U ofthe polymer-shelled microbubbles is approximately 120 timeshigher than that of the polymer microbeads. The correspondingradiation force acting on the polymer-shelled microbubbles isapproximately 146 times higher than the radiation force actingon the plain polymer microbeads.

The forces acting on the polymer-shelled microbubbles in thetransverse direction are much stronger than those evident for theplain polymer particles, as illustrated in Fig. 7. The transverse com-ponent is very prominent, while the axial motion is almost belowthe threshold for plain particles. The transverse component,because of the non-uniform standing-wave along the channel, pro-duced a 2-D standing wave [48], although we intended to producea 1-D standing wave across the channel. Moreover, both the inter-action between particles and the particle–wall interaction aremuch more prominent for microbubbles, a phenomenon that isnot accounted for in this study (the Gor’kov model assumes onlysingle-particle motion in a 1-D standing wave and does notaccount for other effects, such as a 2-D wave or wall interactions).

The current study is limited to no-flow conditions that is, theparticles are free from flow-related effects, such as microstream-ing. As a result, the buoyancy forces exerted by the fluid on thepolymer-shelled microbubbles bring them to the upper surface ofthe channel, while the same forces exerted on the microbeadsmoved them to the chip’s bottom plane (Fig. 11).

5.1. Possible clinical applications

Recent advances in surface chemistrymake it possible to modifythe encapsulated shell in order to anchor ligands and selectively

target the diseased tissue or organ. The bubbles are today pureblood-pool contrast media and the possible targeting cells areblood-, endothelial-, and macrophage-like cells hiding in the liverand the spleen [7,50]. Reported in this article polymer-shelledmicrobubbles can easily be decorated with dyes [37], ligands (suchas RGD peptides [51]), and antibodies [52] in order to target theendothelial cells and liver inflammationwhere the affinity betweenfunctionalized microbubbles and targeted cells is crucial. Thesepolymer-shelled microbubbles can be potentially utilized in bioan-alytical applications. For example, the functionalized microbubblesbounded to the cells can be potentially trapped at the pressureantinode at a pressure level as low as 60 kPa, leaving free cellsunaffected.

The polymer-shelled microbubbles exhibited stable both linearand nonlinear oscillations below 1 MPa. Moreover, microbubblescan be disrupted without forming secondary free gas bubbles athigh acoustic pressure of about 1 MPa [47], in contrast to thin-shelled and even other types of polymer-shelled microbubbles[53]. This disruption process is attractive for reversible sonopora-tion applications [54]. In molecular and genetic disease, an acous-tophoretic single-cell lysis characterization has been suggested[55]. The acoustic lysis procedure requires long (approximately50 s) high-pressure ultrasound exposures, resulting in significantheating and denaturation of intracellular components [55]. In thepresence of microbubbles, the acoustic cavitation pressure thresh-old is reduced, as is the exposure time needed.

6. Conclusion

The key knowledge this paper presents is insight into the phys-ical properties and behavior of stable polymer-shelled microsizegas bubbles in acoustic standing-wave fields of low megahertz fre-quency of 2.8 MHz and pressure below 165 kPa. Theoretical con-siderations and the calculated acoustic contrast factor(U = �60.7) hinted that the primary radiation force drags thepolymer-shelled microbubbles to the pressure antinode planes;this was experimentally confirmed. The polymer-shelledmicrobubbles translated to the pressure antinode planes withinhalf a second, corresponding to an acoustic radiation force 146times higher than that for plain polymer beads, Moreover, thesemicrobubbles could be separated from solid particles. The high

S.V.V.N. Kothapalli et al. / Ultrasonics 70 (2016) 275–283 283

negative U of polymer-shelled microbubbles can be beneficial inperforming bioanalytical applications in vitro with the help ofacoustophoresis methods. This experimental evidence of stabilizedgas bubbles’ behavior in an acoustic standing-wave resonator canbe further extended to cell sorting, cell lysis, or bioaffinity studies.

Acknowledgements

Authors would like to appreciate Muhammed Asim Faridhhi,Mathias Ohlin and Ida Sadat Iranmanesh for their thorough supportand enthusiastic discussions. The study was partly covered by EU-grants (3MiCRON) and strategic money from Karolinska Institute.

References

[1] J. Ophir, K.J. Parker, Contrast agents in diagnostic ultrasound, Ultrasound Med.Biol. 15 (1989) 319–333.

[2] R. Gramiak, P.M. Shah, Echocardiography of the aortic root, Invest. Radiol. 3(1968) 356–366.

[3] S.B. Feinstein, P.M. Shah, R.J. Bing, S. Meerbaum, E. Corday, B.-L. Chang, G.Santillan, Y. Fujibayashi, Microbubble dynamics visualized in the intactcapillary circulation, J. Am. Coll. Cardiol. 4 (1984) 595–600.

[4] M. Schneider, SonoVue, a new ultrasound contrast agent, Eur. Radiol. 9 (1999)S347.

[5] M.A. Wheatley, B. Schrope, P. Shen, Contrast agents for diagnostic ultrasound:development and evaluation of polymer-coated microbubbles, Biomaterials 11(1990) 713–717.

[6] A.L. Klibanov, Microbubble contrast agents: targeted ultrasound imaging andultrasound-assisted drug-delivery applications, Invest. Radiol. 41 (2006) 354–362.

[7] K. Ferrara, R. Pollard, M. Borden, Ultrasound microbubble contrast agents:fundamentals and application to gene and drug delivery, Biomed. Eng. 9(2007).

[8] J.R. Lindner, J. Song, J. Christiansen, A.L. Klibanov, F. Xu, K. Ley, Ultrasoundassessment of inflammation and renal tissue injury with microbubblestargeted to P-selectin, Circulation 104 (2001) 2107–2112.

[9] J.J. Rychak, A.L. Klibanov, K.F. Ley, J.A. Hossack, Enhanced targeting ofultrasound contrast agents using acoustic radiation force, Ultrasound Med.Biol. 33 (2007) 1132–1139.

[10] P.A. Dayton, J.S. Allen, K.W. Ferrara, The magnitude of radiation force onultrasound contrast agents, J. Acoust. Soc. Am. 112 (2002) 2183–2192.

[11] L. Rayleigh, XXXIV. On the pressure of vibrations, Lond., Edinburgh, DublinPhilos. Mag. J. Sci. 3 (1902) 338–346.

[12] M. Fatemi, J.F. Greenleaf, Vibro-acoustography: an imaging modality based onultrasound-stimulated acoustic emission, Proc. Natl. Acad. Sci. USA 96 (1999)6603–6608.

[13] F. Sebag, J. Vaillant-Lombard, J. Berbis, V. Griset, J. Henry, P. Petit, C. Oliver,Shear wave elastography: a new ultrasound imaging mode for the differentialdiagnosis of benign and malignant thyroid nodules, J. Clin. Endocrinol. Metab.95 (2010) 5281–5288.

[14] K. Nightingale, M.S. Soo, R. Nightingale, G. Trahey, Acoustic radiation forceimpulse imaging: in vivo demonstration of clinical feasibility, Ultrasound Med.Biol. 28 (2002) 227–235.

[15] R. Muthupillai, D. Lomas, P. Rossman, J. Greenleaf, A. Manduca, R. Ehman,Magnetic resonance elastography by direct visualization of propagatingacoustic strain waves, Science 269 (1995) 1854–1857.

[16] W.F. Walker, F.J. Fernandez, L.A. Negron, A method of imaging viscoelasticparameters with acoustic radiation force, Phys. Med. Biol. 45 (2000) 1437.

[17] W. Coakley, G. Whitworth, M. Grundy, R. Gould, R. Allman, Ultrasonicmanipulation of particles and cells. Ultrasonic separation of cells,Bioseparation 4 (1994) 73–83.

[18] F. Blake Jr., Bjerknes forces in stationary sound fields, J. Acoust. Soc. Am. 21(1949) 551.

[19] L.A. Crum, Bjerknes forces on bubbles in a stationary sound field, J. Acoust. Soc.Am. 57 (1975) 1363–1370.

[20] A.A. Doinikov, Translational motion of a spherical bubble in an acousticstanding wave of high intensity, Phys. Fluids 14 (1994–present) 1420–1425.

[21] A. Kundt, O. Lehmann, Über longitudinale Schwingungen and Klangfiguren inzylindrischen Flüssigkeitssäulen, Ann. Phys. 135 (1874) 1–12.

[22] L. Gor’Kov, On the forces acting on a small particle in an acoustical field in anideal fluid, in: Soviet Physics Doklady, 1962, p. 773.

[23] K. Yosioka, Y. Kawasima, Acoustic radiation pressure on a compressible sphere,Acta Acust. United Acust. 5 (1955) 167–173.

[24] A. Lenshof, T. Laurell, Emerging clinical applications of microchip-basedacoustophoresis, J. Assoc. Lab. Autom. 16 (2011) 443–449.

[25] A. Lenshof, C. Magnusson, T. Laurell, Acoustofluidics 8: applications ofacoustophoresis in continuous flow microsystems, Lab Chip 12 (2012) 1210–1223.

[26] M. Grösch, W. Burger, B. Handl, O. Doblhoff-Dier, T. Gaida, C. Schmatz,Ultrasonic separation of suspended particles – Part III: Application inbiotechnology, Acta Acust. United Acust. 84 (1998) 815–822.

[27] F. Petersson, A. Nilsson, C. Holm, H. Jönsson, T. Laurell, Separation of lipidsfrom blood utilizing ultrasonic standing waves in microfluidic channels,Analyst 129 (2004) 938–943.

[28] F. Petersson, L. Åberg, A.-M. Swärd-Nilsson, T. Laurell, Free flowacoustophoresis: microfluidic-based mode of particle and cell separation,Anal. Chem. 79 (2007) 5117–5123.

[29] C. Grenvall, P. Augustsson, J.R. Folkenberg, T. Laurell, Harmonic microchipacoustophoresis: a route to online raw milk sample precondition in proteinand lipid content quality control, Anal. Chem. 81 (2009) 6195–6200.

[30] H. Wang, Z. Liu, S. Kim, C. Koo, Y. Cho, D.-Y. Jang, Y.-J. Kim, A. Han, Microfluidicacustophoretic force based low-concentration oil separation and detectionfrom the environment, Lab Chip 14 (2014) 947–956.

[31] K.W. Cushing, M.E. Piyasena, N.J. Carroll, G.C. Maestas, B.A. López, B.S. Edwards,S.W. Graves, G.P. López, Elastomeric negative acoustic contrast particles foraffinity capture assays, Anal. Chem. 85 (2013) 2208–2215.

[32] C.W. Shields IV, L.M. Johnson, L. Gao, G.P. López, Elastomeric negative acousticcontrast particles for capture, acoustophoretic transport, and confinement ofcells in microfluidic systems, Langmuir 30 (2014) 3923–3927.

[33] I. Leibacher, W. Dietze, P. Hahn, J. Wang, S. Schmitt, J. Dual, Acoustophoresis ofhollow and core-shell particles in two-dimensional resonance modes,Microfluid. Nanofluid. 16 (2014) 513–524.

[34] L. Hoff, Acoustic Characterization of Contrast Agents for Medical UltrasoundImaging, Springer, 2001.

[35] S. Khanna, N.N. Amso, S.J. Paynter, W.T. Coakley, Contrast agent bubble anderythrocyte behavior in a 1.5-MHz standing ultrasound wave, Ultrasound Med.Biol. 29 (2003) 1463–1470.

[36] D. Grishenkov, C. Pecorari, T.B. Brismar, G. Paradossi, Characterization ofacoustic properties of PVA-shelled ultrasound contrast agents: linearproperties (part I), Ultrasound Med. Biol. 35 (2009) 1127–1138.

[37] F. Cavalieri, A. El Hamassi, E. Chiessi, G. Paradossi, Stable polymericmicroballoons as multifunctional device for biomedical uses: synthesis andcharacterization, Langmuir 21 (2005) 8758–8764.

[38] S. Sirsi, M. Borden, Microbubble compositions, properties and biomedicalapplications, Bubble Sci., Eng. Technol. 1 (2009) 3–17.

[39] D. Grishenkov, A. Gonon, E. Weitzberg, J. Lundberg, J. Harmark, B. Cerroni, G.Paradossi, B. Janerot-Sjoberg, Ultrasound contrast agent loaded with nitricoxide as a theranostic microdevice for myocardial ischemia, J. Drug Des., Dev.Ther. 2015 (9) (2015) 2409–2419.

[40] R. Barnkob, I. Iranmanesh, M. Wiklund, H. Bruus, Measuring acoustic energydensity in microchannel acoustophoresis using a simple and rapid light-intensity method, Lab Chip 12 (2012) 2337–2344.

[41] M. Wiklund, H.M. Hertz, Ultrasonic enhancement of bead-based bioaffinityassays, Lab Chip 6 (2006) 1279–1292.

[42] R. Barnkob, P. Augustsson, T. Laurell, H. Bruus, Measuring the local pressureamplitude in microchannel acoustophoresis, Lab Chip 10 (2010) 563–570.

[43] P.A. Fernandes, G. Tzvetkov, R.H. Fink, G. Paradossi, A. Fery, Quantitativeanalysis of scanning transmission X-ray microscopy images of gas-filled PVA-based microballoons, Langmuir 24 (2008) 13677–13682.

[44] M. Poehlmann, D. Grishenkov, S.V. Kothapalli, J. Härmark, H. Hebert, A. Philipp,R. Hoeller, M. Seuss, C. Kuttner, S. Margheritelli, On the interplay of shellstructure with low-and high-frequency mechanics of multifunctionalmagnetic microbubbles, Soft Matter 10 (2014) 214–226.

[45] M. Wiklund, J. Toivonen, M. Tirri, P. Hänninen, H.M. Hertz, Ultrasonicenrichment of microspheres for ultrasensitive biomedical analysis inconfocal laser-scanning fluorescence detection, J. Appl. Phys. 96 (2004)1242–1248.

[46] C. Sciallero, D. Grishenkov, S. Kothapalli, L. Oddo, A. Trucco, Acousticcharacterization and contrast imaging of microbubbles encapsulated bypolymeric shells coated or filled with magnetic nanoparticles, J. Acoust. Soc.Am. 134 (2013) 3918–3930.

[47] D. Grishenkov, C. Pecorari, T.B. Brismar, G. Paradossi, Characterization ofacoustic properties of PVA-shelled ultrasound contrast agents: linearproperties (part II), Ultrasound Med. Biol. 35 (2009) 1139–1147.

[48] S. Hagsäter, A. Lenshof, P. Skafte-Pedersen, J.P. Kutter, T. Laurell, H. Bruus,Acoustic resonances in straight micro channels: beyond the 1D-approximation, Lab Chip 8 (2008) 1178–1184.

[49] N. De Jong, M. Emmer, A. Van Wamel, M. Versluis, Ultrasonic characterizationof ultrasound contrast agents, Med. Biol. Eng. Comput. 47 (2009) 861–873.

[50] A.L. Klibanov, Targeted delivery of gas-filled microspheres, contrast agents forultrasound imaging, Adv. Drug Deliv. Rev. 37 (1999) 139–157.

[51] M. Tortora, L. Oddo, S. Margheritelli, G. Paradossi, Design of novel polymershelled ultrasound contrast agents: towards an ultrasound triggered drugdelivery, in: Ultrasound Contrast Agents, Springer, 2010, pp. 25–39.

[52] R. Villa, B. Cerroni, L. Viganò, S. Margheritelli, G. Abolafio, L. Oddo, G. Paradossi,N. Zaffaroni, Targeted doxorubicin delivery by chitosan-galactosylatedmodified polymer microbubbles to hepatocarcinoma cells, Colloids Surf., B:Biointerfaces 110 (2013) 434–442.

[53] S.H. Bloch, M. Wan, P.A. Dayton, K.W. Ferrara, Optical observation of lipid-andpolymer-shelled ultrasound microbubble contrast agents, Appl. Phys. Lett. 84(2004) 631–633.

[54] K. Kooiman, M. Foppen-Harteveld, A.F. van der Steen, N. de Jong, Sonoporationof endothelial cells by vibrating targeted microbubbles, J. Control. Release 154(2011) 35–41.

[55] R.B. Brown, J. Audet, Current techniques for single-cell lysis, J. R. Soc. Interface5 (2008) S131–S138.